Face Generation

In this project, you'll use generative adversarial networks to generate new images of faces.

Get the Data

You'll be using two datasets in this project:

  • MNIST
  • CelebA

Since the celebA dataset is complex and you're doing GANs in a project for the first time, we want you to test your neural network on MNIST before CelebA. Running the GANs on MNIST will allow you to see how well your model trains sooner.

If you're using FloydHub, set data_dir to "/input" and use the FloydHub data ID "R5KrjnANiKVhLWAkpXhNBe".

In [1]:
data_dir = './data'

# FloydHub - Use with data ID "R5KrjnANiKVhLWAkpXhNBe"
#data_dir = '/input'


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import helper

helper.download_extract('mnist', data_dir)
helper.download_extract('celeba', data_dir)
Found mnist Data
Found celeba Data

Explore the Data

MNIST

As you're aware, the MNIST dataset contains images of handwritten digits. You can view the first number of examples by changing show_n_images.

In [2]:
show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
%matplotlib inline
import os
from glob import glob
from matplotlib import pyplot

mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'mnist/*.jpg'))[:show_n_images], 28, 28, 'L')
pyplot.imshow(helper.images_square_grid(mnist_images, 'L'), cmap='gray')
Out[2]:
<matplotlib.image.AxesImage at 0x7f895a65a390>

CelebA

The CelebFaces Attributes Dataset (CelebA) dataset contains over 200,000 celebrity images with annotations. Since you're going to be generating faces, you won't need the annotations. You can view the first number of examples by changing show_n_images.

In [3]:
show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'img_align_celeba/*.jpg'))[:show_n_images], 28, 28, 'RGB')
pyplot.imshow(helper.images_square_grid(mnist_images, 'RGB'))
Out[3]:
<matplotlib.image.AxesImage at 0x7f895a57ef60>

Preprocess the Data

Since the project's main focus is on building the GANs, we'll preprocess the data for you. The values of the MNIST and CelebA dataset will be in the range of -0.5 to 0.5 of 28x28 dimensional images. The CelebA images will be cropped to remove parts of the image that don't include a face, then resized down to 28x28.

The MNIST images are black and white images with a single color channel while the CelebA images have 3 color channels (RGB color channel).

Build the Neural Network

You'll build the components necessary to build a GANs by implementing the following functions below:

  • model_inputs
  • discriminator
  • generator
  • model_loss
  • model_opt
  • train

Check the Version of TensorFlow and Access to GPU

This will check to make sure you have the correct version of TensorFlow and access to a GPU

In [4]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from distutils.version import LooseVersion
import warnings
import tensorflow as tf

# Check TensorFlow Version
assert LooseVersion(tf.__version__) >= LooseVersion('1.0'), 'Please use TensorFlow version 1.0 or newer.  You are using {}'.format(tf.__version__)
print('TensorFlow Version: {}'.format(tf.__version__))

# Check for a GPU
if not tf.test.gpu_device_name():
    warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
    print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))
TensorFlow Version: 1.2.1
Default GPU Device: /gpu:0

Input

Implement the model_inputs function to create TF Placeholders for the Neural Network. It should create the following placeholders:

  • Real input images placeholder with rank 4 using image_width, image_height, and image_channels.
  • Z input placeholder with rank 2 using z_dim.
  • Learning rate placeholder with rank 0.

Return the placeholders in the following the tuple (tensor of real input images, tensor of z data)

In [5]:
import problem_unittests as tests

def model_inputs(image_width, image_height, image_channels, z_dim):
    """
    Create the model inputs
    :param image_width: The input image width
    :param image_height: The input image height
    :param image_channels: The number of image channels
    :param z_dim: The dimension of Z
    :return: Tuple of (tensor of real input images, tensor of z data, learning rate)
    """
    # TODO: Implement Function
    input_real = tf.placeholder(tf.float32, shape=[None, image_width, image_height, image_channels], name='input_real')
    input_z = tf.placeholder(tf.float32, shape=[None, z_dim], name='input_z')
    learning_rate = tf.placeholder(tf.float32, name='learning_rate')
    return input_real, input_z, learning_rate

"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_inputs(model_inputs)
ERROR:tensorflow:==================================
Object was never used (type <class 'tensorflow.python.framework.ops.Operation'>):
<tf.Operation 'assert_rank_2/Assert/Assert' type=Assert>
If you want to mark it as used call its "mark_used()" method.
It was originally created here:
['File "/home/carnd/anaconda3/envs/dl/lib/python3.5/runpy.py", line 184, in _run_module_as_main\n    "__main__", mod_spec)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/runpy.py", line 85, in _run_code\n    exec(code, run_globals)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/ipykernel/__main__.py", line 3, in <module>\n    app.launch_new_instance()', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/traitlets/config/application.py", line 658, in launch_instance\n    app.start()', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/ipykernel/kernelapp.py", line 474, in start\n    ioloop.IOLoop.instance().start()', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/zmq/eventloop/ioloop.py", line 177, in start\n    super(ZMQIOLoop, self).start()', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tornado/ioloop.py", line 887, in start\n    handler_func(fd_obj, events)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tornado/stack_context.py", line 275, in null_wrapper\n    return fn(*args, **kwargs)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/zmq/eventloop/zmqstream.py", line 440, in _handle_events\n    self._handle_recv()', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/zmq/eventloop/zmqstream.py", line 472, in _handle_recv\n    self._run_callback(callback, msg)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/zmq/eventloop/zmqstream.py", line 414, in _run_callback\n    callback(*args, **kwargs)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tornado/stack_context.py", line 275, in null_wrapper\n    return fn(*args, **kwargs)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/ipykernel/kernelbase.py", line 276, in dispatcher\n    return self.dispatch_shell(stream, msg)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/ipykernel/kernelbase.py", line 228, in dispatch_shell\n    handler(stream, idents, msg)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/ipykernel/kernelbase.py", line 390, in execute_request\n    user_expressions, allow_stdin)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/ipykernel/ipkernel.py", line 196, in do_execute\n    res = shell.run_cell(code, store_history=store_history, silent=silent)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/ipykernel/zmqshell.py", line 501, in run_cell\n    return super(ZMQInteractiveShell, self).run_cell(*args, **kwargs)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/IPython/core/interactiveshell.py", line 2717, in run_cell\n    interactivity=interactivity, compiler=compiler, result=result)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/IPython/core/interactiveshell.py", line 2827, in run_ast_nodes\n    if self.run_code(code, result):', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/IPython/core/interactiveshell.py", line 2881, in run_code\n    exec(code_obj, self.user_global_ns, self.user_ns)', 'File "<ipython-input-5-87d120f90811>", line 21, in <module>\n    tests.test_model_inputs(model_inputs)', 'File "/home/carnd/deep-learning/face_generation/problem_unittests.py", line 12, in func_wrapper\n    result = func(*args)', 'File "/home/carnd/deep-learning/face_generation/problem_unittests.py", line 68, in test_model_inputs\n    _check_input(learn_rate, [], \'Learning Rate\')', 'File "/home/carnd/deep-learning/face_generation/problem_unittests.py", line 34, in _check_input\n    _assert_tensor_shape(tensor, shape, \'Real Input\')', 'File "/home/carnd/deep-learning/face_generation/problem_unittests.py", line 20, in _assert_tensor_shape\n    assert tf.assert_rank(tensor, len(shape), message=\'{} has wrong rank\'.format(display_name))', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tensorflow/python/ops/check_ops.py", line 617, in assert_rank\n    dynamic_condition, data, summarize)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tensorflow/python/ops/check_ops.py", line 571, in _assert_rank_condition\n    return control_flow_ops.Assert(condition, data, summarize=summarize)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tensorflow/python/util/tf_should_use.py", line 170, in wrapped\n    return _add_should_use_warning(fn(*args, **kwargs))', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tensorflow/python/util/tf_should_use.py", line 139, in _add_should_use_warning\n    wrapped = TFShouldUseWarningWrapper(x)', 'File "/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tensorflow/python/util/tf_should_use.py", line 96, in __init__\n    stack = [s.strip() for s in traceback.format_stack()]']
==================================
Tests Passed

Discriminator

Implement discriminator to create a discriminator neural network that discriminates on images. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "discriminator" to allow the variables to be reused. The function should return a tuple of (tensor output of the discriminator, tensor logits of the discriminator).

In [8]:
def discriminator(images, reuse=False):
    """
    Create the discriminator network
    :param images: Tensor of input image(s)
    :param reuse: Boolean if the weights should be reused
    :return: Tuple of (tensor output of the discriminator, tensor logits of the discriminator)
    """
    # TODO: Implement Function
    keep_prob = 0.5
    alpha = 0.2

    with tf.variable_scope('discriminator', reuse=reuse):
        
        layer1 = tf.layers.conv2d(images, 64, 5, strides=2, padding='same')
        layer1 = tf.layers.dropout(layer1, rate=(keep_prob), training=True)
        batch_norm_1 = tf.layers.batch_normalization(layer1, training=True)
        relu1 = tf.maximum(alpha * layer1, layer1)
        
        layer2 = tf.layers.conv2d(relu1, 128, 5, strides=2, padding='same')
        layer2 = tf.layers.dropout(layer2, rate=(keep_prob), training=True)
        batch_norm_2 = tf.layers.batch_normalization(layer2, training=True)
        relu2 = tf.maximum(alpha * batch_norm_2, batch_norm_2)
        
        layer3 = tf.layers.conv2d(relu2, 256, 5, strides=2, padding='same')
        layer3 = tf.layers.dropout(layer3, rate=(keep_prob), training=True)
        batch_norm_3 = tf.layers.batch_normalization(layer3, training=True)
        relu3 = tf.maximum(alpha * batch_norm_3, batch_norm_3)

        flat = tf.reshape(relu3, (-1, 4*4*256))
        logits = tf.layers.dense(flat, 1)
        out = tf.sigmoid(logits)
        
        return out, logits

"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_discriminator(discriminator, tf)
Tests Passed

Generator

Implement generator to generate an image using z. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "generator" to allow the variables to be reused. The function should return the generated 28 x 28 x out_channel_dim images.

In [9]:
def generator(z, out_channel_dim, is_train=True):
    """
    Create the generator network
    :param z: Input z
    :param out_channel_dim: The number of channels in the output image
    :param is_train: Boolean if generator is being used for training
    :return: The tensor output of the generator
    """
    # TODO: Implement Function
    
    alpha = 0.1
    keep_prob = 0.5
    
    with tf.variable_scope("generator", reuse=not is_train):
        
        x = tf.layers.dense(z, 7*7*512)
        x = tf.reshape(x, (-1, 7, 7, 512))
        x = tf.layers.batch_normalization(x, training=is_train)
        x = tf.maximum(alpha * x, x)
        
        layer1 = tf.layers.conv2d_transpose(x, 256, 5, strides=1, padding="same")
        layer1 = tf.layers.dropout(layer1, rate=(keep_prob), training=is_train)
        layer1 = tf.layers.batch_normalization(layer1, training=is_train)
        layer1 = tf.maximum(alpha * layer1, layer1)
        
        layer2 = tf.layers.conv2d_transpose(layer1, 128, 5, strides=2, padding="same")
        layer2 = tf.layers.dropout(layer2, rate=(keep_prob), training=is_train)
        layer2 = tf.layers.batch_normalization(layer2, training=is_train)
        layer2 = tf.maximum(alpha * layer2, layer2)
        
        logits = tf.layers.conv2d_transpose(layer2, out_channel_dim, 5, strides=2, padding="same")
        out = tf.tanh(logits)
    
    return out

"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_generator(generator, tf)
Tests Passed

Loss

Implement model_loss to build the GANs for training and calculate the loss. The function should return a tuple of (discriminator loss, generator loss). Use the following functions you implemented:

  • discriminator(images, reuse=False)
  • generator(z, out_channel_dim, is_train=True)
In [11]:
def model_loss(input_real, input_z, out_channel_dim):
    """
    Get the loss for the discriminator and generator
    :param input_real: Images from the real dataset
    :param input_z: Z input
    :param out_channel_dim: The number of channels in the output image
    :return: A tuple of (discriminator loss, generator loss)
    """
    # TODO: Implement Function
    
    g_model = generator(input_z, out_channel_dim)
    d_model_real, d_logits_real = discriminator(input_real)
    d_model_fake, d_logits_fake = discriminator(g_model, reuse=True)
    
    smooth = 0.1

    d_loss_real = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_real, labels=tf.ones_like(d_model_real) * (1 - smooth)))
    d_loss_fake = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.zeros_like(d_model_fake)))
    g_loss = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.ones_like(d_model_fake)))

    d_loss = d_loss_real + d_loss_fake
    
    return d_loss, g_loss


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_loss(model_loss)
Tests Passed

Optimization

Implement model_opt to create the optimization operations for the GANs. Use tf.trainable_variables to get all the trainable variables. Filter the variables with names that are in the discriminator and generator scope names. The function should return a tuple of (discriminator training operation, generator training operation).

In [12]:
def model_opt(d_loss, g_loss, learning_rate, beta1):
    """
    Get optimization operations
    :param d_loss: Discriminator loss Tensor
    :param g_loss: Generator loss Tensor
    :param learning_rate: Learning Rate Placeholder
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :return: A tuple of (discriminator training operation, generator training operation)
    """
    # TODO: Implement Function
    
    t_vars = tf.trainable_variables()
    d_vars = [var for var in t_vars if var.name.startswith('discriminator')]
    g_vars = [var for var in t_vars if var.name.startswith('generator')]

    # Optimize
    with tf.control_dependencies(tf.get_collection(tf.GraphKeys.UPDATE_OPS)):
        d_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(d_loss, var_list=d_vars)
        g_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(g_loss, var_list=g_vars)

    return d_train_opt, g_train_opt


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_opt(model_opt, tf)
Tests Passed

Neural Network Training

Show Output

Use this function to show the current output of the generator during training. It will help you determine how well the GANs is training.

In [13]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np

def show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode):
    """
    Show example output for the generator
    :param sess: TensorFlow session
    :param n_images: Number of Images to display
    :param input_z: Input Z Tensor
    :param out_channel_dim: The number of channels in the output image
    :param image_mode: The mode to use for images ("RGB" or "L")
    """
    cmap = None if image_mode == 'RGB' else 'gray'
    z_dim = input_z.get_shape().as_list()[-1]
    example_z = np.random.uniform(-1, 1, size=[n_images, z_dim])

    samples = sess.run(
        generator(input_z, out_channel_dim, False),
        feed_dict={input_z: example_z})

    images_grid = helper.images_square_grid(samples, image_mode)
    pyplot.imshow(images_grid, cmap=cmap)
    pyplot.show()

Train

Implement train to build and train the GANs. Use the following functions you implemented:

  • model_inputs(image_width, image_height, image_channels, z_dim)
  • model_loss(input_real, input_z, out_channel_dim)
  • model_opt(d_loss, g_loss, learning_rate, beta1)

Use the show_generator_output to show generator output while you train. Running show_generator_output for every batch will drastically increase training time and increase the size of the notebook. It's recommended to print the generator output every 100 batches.

In [19]:
def train(epoch_count, batch_size, z_dim, learning_rate, beta1, get_batches, data_shape, data_image_mode):
    """
    Train the GAN
    :param epoch_count: Number of epochs
    :param batch_size: Batch Size
    :param z_dim: Z dimension
    :param learning_rate: Learning Rate
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :param get_batches: Function to get batches
    :param data_shape: Shape of the data
    :param data_image_mode: The image mode to use for images ("RGB" or "L")
    """
    # TODO: Build Model
    
    input_real, input_z, learningrate = model_inputs(data_shape[1], data_shape[2], data_shape[3], z_dim)
    
    d_loss, g_loss = model_loss(input_real, input_z, data_shape[3])
    d_opt, g_opt = model_opt(d_loss, g_loss, learningrate, beta1)
    
    steps = 0
    show_every=100
    
    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())
        for epoch_i in range(epoch_count):
            for batch_images in get_batches(batch_size):
                
                # TODO: Train Model
                steps += 1
                batch_images *= 2
                batch_z = np.random.uniform(-1, 1, size=(batch_size, z_dim))

                # Optimizers
                _ = sess.run(d_opt, feed_dict={input_real: batch_images, input_z: batch_z, learningrate: learning_rate})
                _ = sess.run(g_opt, feed_dict={input_z: batch_z, input_real: batch_images, learningrate: learning_rate})

                if steps % show_every == 0:
                    # At the end of each epoch, get the losses and print them out
                    train_loss_d = d_loss.eval({input_z: batch_z, input_real: batch_images})
                    train_loss_g = g_loss.eval({input_z: batch_z})

                    print("Epoch {}/{} Step {}...".format(epoch_i+1, epoch_count, steps),
                          "Discriminator Loss: {:.4f}...".format(train_loss_d),
                          "Generator Loss: {:.4f}".format(train_loss_g))

                    show_generator_output(sess, 25, input_z, data_shape[3], data_image_mode)
    
                

MNIST

Test your GANs architecture on MNIST. After 2 epochs, the GANs should be able to generate images that look like handwritten digits. Make sure the loss of the generator is lower than the loss of the discriminator or close to 0.

In [20]:
batch_size = 32
z_dim = 100
learning_rate = 0.0005
beta1 = 0.5

"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 2

mnist_dataset = helper.Dataset('mnist', glob(os.path.join(data_dir, 'mnist/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, mnist_dataset.get_batches,
          mnist_dataset.shape, mnist_dataset.image_mode)
Epoch 1/2 Step 100... Discriminator Loss: 1.4692... Generator Loss: 1.1016
Epoch 1/2 Step 200... Discriminator Loss: 1.6123... Generator Loss: 0.7882
Epoch 1/2 Step 300... Discriminator Loss: 1.0774... Generator Loss: 1.3213
Epoch 1/2 Step 400... Discriminator Loss: 1.4253... Generator Loss: 0.7557
Epoch 1/2 Step 500... Discriminator Loss: 1.0381... Generator Loss: 1.4679
Epoch 1/2 Step 600... Discriminator Loss: 1.3500... Generator Loss: 1.1109
Epoch 1/2 Step 700... Discriminator Loss: 1.1638... Generator Loss: 0.9626
Epoch 1/2 Step 800... Discriminator Loss: 1.2989... Generator Loss: 0.8193
Epoch 1/2 Step 900... Discriminator Loss: 1.1421... Generator Loss: 1.1247
Epoch 1/2 Step 1000... Discriminator Loss: 1.2368... Generator Loss: 0.8832
Epoch 1/2 Step 1100... Discriminator Loss: 1.2415... Generator Loss: 1.0446
Epoch 1/2 Step 1200... Discriminator Loss: 1.2793... Generator Loss: 1.0336
Epoch 1/2 Step 1300... Discriminator Loss: 1.1497... Generator Loss: 1.0073
Epoch 1/2 Step 1400... Discriminator Loss: 1.3801... Generator Loss: 0.8769
Epoch 1/2 Step 1500... Discriminator Loss: 1.4033... Generator Loss: 0.6571
Epoch 1/2 Step 1600... Discriminator Loss: 1.1625... Generator Loss: 0.7577
Epoch 1/2 Step 1700... Discriminator Loss: 1.3764... Generator Loss: 0.7666
Epoch 1/2 Step 1800... Discriminator Loss: 1.1899... Generator Loss: 1.0114
Epoch 2/2 Step 1900... Discriminator Loss: 1.2471... Generator Loss: 0.9072
Epoch 2/2 Step 2000... Discriminator Loss: 1.1407... Generator Loss: 1.5529
Epoch 2/2 Step 2100... Discriminator Loss: 1.1523... Generator Loss: 0.8752
Epoch 2/2 Step 2200... Discriminator Loss: 1.1636... Generator Loss: 0.7601
Epoch 2/2 Step 2300... Discriminator Loss: 1.2331... Generator Loss: 0.8365
Epoch 2/2 Step 2400... Discriminator Loss: 1.5334... Generator Loss: 0.7301
Epoch 2/2 Step 2500... Discriminator Loss: 1.4137... Generator Loss: 0.7020
Epoch 2/2 Step 2600... Discriminator Loss: 1.2270... Generator Loss: 1.0844
Epoch 2/2 Step 2700... Discriminator Loss: 1.1106... Generator Loss: 1.1797
Epoch 2/2 Step 2800... Discriminator Loss: 1.0402... Generator Loss: 1.1101
Epoch 2/2 Step 2900... Discriminator Loss: 1.0964... Generator Loss: 0.8492
Epoch 2/2 Step 3000... Discriminator Loss: 1.0731... Generator Loss: 1.3411
Epoch 2/2 Step 3100... Discriminator Loss: 1.1684... Generator Loss: 0.8948
Epoch 2/2 Step 3200... Discriminator Loss: 1.1624... Generator Loss: 1.3229
Epoch 2/2 Step 3300... Discriminator Loss: 1.1274... Generator Loss: 0.6103
Epoch 2/2 Step 3400... Discriminator Loss: 1.2848... Generator Loss: 1.1488
Epoch 2/2 Step 3500... Discriminator Loss: 1.0138... Generator Loss: 1.1882
Epoch 2/2 Step 3600... Discriminator Loss: 1.0518... Generator Loss: 1.2696
Epoch 2/2 Step 3700... Discriminator Loss: 0.8857... Generator Loss: 1.3543

CelebA

Run your GANs on CelebA. It will take around 20 minutes on the average GPU to run one epoch. You can run the whole epoch or stop when it starts to generate realistic faces.

In [21]:
batch_size = 32
z_dim = 100
learning_rate = 0.0005
beta1 = 0.5

"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 1

celeba_dataset = helper.Dataset('celeba', glob(os.path.join(data_dir, 'img_align_celeba/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, celeba_dataset.get_batches,
          celeba_dataset.shape, celeba_dataset.image_mode)
Epoch 1/1 Step 100... Discriminator Loss: 1.0732... Generator Loss: 1.7619
Epoch 1/1 Step 200... Discriminator Loss: 0.7502... Generator Loss: 1.5938
Epoch 1/1 Step 300... Discriminator Loss: 0.7843... Generator Loss: 1.6228
Epoch 1/1 Step 400... Discriminator Loss: 0.6167... Generator Loss: 2.2576
Epoch 1/1 Step 500... Discriminator Loss: 1.3154... Generator Loss: 0.9045
Epoch 1/1 Step 600... Discriminator Loss: 1.2655... Generator Loss: 1.3770
Epoch 1/1 Step 700... Discriminator Loss: 0.9746... Generator Loss: 2.1074
Epoch 1/1 Step 800... Discriminator Loss: 1.1541... Generator Loss: 0.9664
Epoch 1/1 Step 900... Discriminator Loss: 1.1324... Generator Loss: 1.2113
Epoch 1/1 Step 1000... Discriminator Loss: 1.0265... Generator Loss: 0.9492
Epoch 1/1 Step 1100... Discriminator Loss: 0.9424... Generator Loss: 1.3833
Epoch 1/1 Step 1200... Discriminator Loss: 1.3368... Generator Loss: 1.7483
Epoch 1/1 Step 1300... Discriminator Loss: 0.9608... Generator Loss: 1.2971
Epoch 1/1 Step 1400... Discriminator Loss: 0.9522... Generator Loss: 1.5322
Epoch 1/1 Step 1500... Discriminator Loss: 1.3128... Generator Loss: 0.9773
Epoch 1/1 Step 1600... Discriminator Loss: 1.3444... Generator Loss: 1.0677
Epoch 1/1 Step 1700... Discriminator Loss: 0.9648... Generator Loss: 1.4728
Epoch 1/1 Step 1800... Discriminator Loss: 1.3306... Generator Loss: 1.0158
Epoch 1/1 Step 1900... Discriminator Loss: 1.2755... Generator Loss: 0.9968
Epoch 1/1 Step 2000... Discriminator Loss: 1.2196... Generator Loss: 1.0400
Epoch 1/1 Step 2100... Discriminator Loss: 1.2497... Generator Loss: 1.0197
Epoch 1/1 Step 2200... Discriminator Loss: 1.2044... Generator Loss: 1.0995
Epoch 1/1 Step 2300... Discriminator Loss: 1.3306... Generator Loss: 0.9016
Epoch 1/1 Step 2400... Discriminator Loss: 1.2917... Generator Loss: 0.8233
Epoch 1/1 Step 2500... Discriminator Loss: 1.1966... Generator Loss: 1.1442
Epoch 1/1 Step 2600... Discriminator Loss: 1.3911... Generator Loss: 1.3014
Epoch 1/1 Step 2700... Discriminator Loss: 1.2063... Generator Loss: 1.3891
Epoch 1/1 Step 2800... Discriminator Loss: 1.2906... Generator Loss: 0.9009
Epoch 1/1 Step 2900... Discriminator Loss: 1.2146... Generator Loss: 0.8924
Epoch 1/1 Step 3000... Discriminator Loss: 1.0883... Generator Loss: 0.9375
Epoch 1/1 Step 3100... Discriminator Loss: 1.1574... Generator Loss: 0.8406
Epoch 1/1 Step 3200... Discriminator Loss: 1.4610... Generator Loss: 0.9256
Epoch 1/1 Step 3300... Discriminator Loss: 1.2512... Generator Loss: 0.9960
Epoch 1/1 Step 3400... Discriminator Loss: 1.3339... Generator Loss: 1.0524
Epoch 1/1 Step 3500... Discriminator Loss: 1.2228... Generator Loss: 0.8074
Epoch 1/1 Step 3600... Discriminator Loss: 1.0758... Generator Loss: 0.9317
Epoch 1/1 Step 3700... Discriminator Loss: 1.1849... Generator Loss: 0.9115
Epoch 1/1 Step 3800... Discriminator Loss: 1.3558... Generator Loss: 0.8512
Epoch 1/1 Step 3900... Discriminator Loss: 1.2718... Generator Loss: 0.9658
Epoch 1/1 Step 4000... Discriminator Loss: 1.2282... Generator Loss: 1.0438
Epoch 1/1 Step 4100... Discriminator Loss: 1.2533... Generator Loss: 1.0073
Epoch 1/1 Step 4200... Discriminator Loss: 1.2000... Generator Loss: 0.8983
Epoch 1/1 Step 4300... Discriminator Loss: 1.4791... Generator Loss: 0.7256
Epoch 1/1 Step 4400... Discriminator Loss: 1.1925... Generator Loss: 1.1429
Epoch 1/1 Step 4500... Discriminator Loss: 1.3415... Generator Loss: 0.9013
Epoch 1/1 Step 4600... Discriminator Loss: 1.3638... Generator Loss: 0.8228
Epoch 1/1 Step 4700... Discriminator Loss: 1.4442... Generator Loss: 0.8557
---------------------------------------------------------------------------
KeyboardInterrupt                         Traceback (most recent call last)
<ipython-input-21-7c85b3eb57ca> in <module>()
     12 with tf.Graph().as_default():
     13     train(epochs, batch_size, z_dim, learning_rate, beta1, celeba_dataset.get_batches,
---> 14           celeba_dataset.shape, celeba_dataset.image_mode)

<ipython-input-19-c93151935ed4> in train(epoch_count, batch_size, z_dim, learning_rate, beta1, get_batches, data_shape, data_image_mode)
     32 
     33                 # Optimizers
---> 34                 _ = sess.run(d_opt, feed_dict={input_real: batch_images, input_z: batch_z, learningrate: learning_rate})
     35                 _ = sess.run(g_opt, feed_dict={input_z: batch_z, input_real: batch_images, learningrate: learning_rate})
     36 

/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tensorflow/python/client/session.py in run(self, fetches, feed_dict, options, run_metadata)
    787     try:
    788       result = self._run(None, fetches, feed_dict, options_ptr,
--> 789                          run_metadata_ptr)
    790       if run_metadata:
    791         proto_data = tf_session.TF_GetBuffer(run_metadata_ptr)

/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tensorflow/python/client/session.py in _run(self, handle, fetches, feed_dict, options, run_metadata)
    995     if final_fetches or final_targets:
    996       results = self._do_run(handle, final_targets, final_fetches,
--> 997                              feed_dict_string, options, run_metadata)
    998     else:
    999       results = []

/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tensorflow/python/client/session.py in _do_run(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)
   1130     if handle is None:
   1131       return self._do_call(_run_fn, self._session, feed_dict, fetch_list,
-> 1132                            target_list, options, run_metadata)
   1133     else:
   1134       return self._do_call(_prun_fn, self._session, handle, feed_dict,

/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tensorflow/python/client/session.py in _do_call(self, fn, *args)
   1137   def _do_call(self, fn, *args):
   1138     try:
-> 1139       return fn(*args)
   1140     except errors.OpError as e:
   1141       message = compat.as_text(e.message)

/home/carnd/anaconda3/envs/dl/lib/python3.5/site-packages/tensorflow/python/client/session.py in _run_fn(session, feed_dict, fetch_list, target_list, options, run_metadata)
   1119         return tf_session.TF_Run(session, options,
   1120                                  feed_dict, fetch_list, target_list,
-> 1121                                  status, run_metadata)
   1122 
   1123     def _prun_fn(session, handle, feed_dict, fetch_list):

KeyboardInterrupt: 

Submitting This Project

When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as "dlnd_face_generation.ipynb" and save it as a HTML file under "File" -> "Download as". Include the "helper.py" and "problem_unittests.py" files in your submission.